English
Related papers

Related papers: Posterior Collapse as Automatic Spectral Pruning

200 papers

We investigate the phenomenon of posterior collapse in variational autoencoders (VAEs) from the perspective of statistical physics, and reveal that it constitutes a phase transition governed jointly by data structure and model…

Machine Learning · Computer Science 2025-12-25 Zhen Li , Fan Zhang , Zheng Zhang , Yu Chen

Due to the phenomenon of "posterior collapse," current latent variable generative models pose a challenging design choice that either weakens the capacity of the decoder or requires augmenting the objective so it does not only maximize the…

Machine Learning · Computer Science 2019-01-14 Ali Razavi , Aäron van den Oord , Ben Poole , Oriol Vinyals

Posterior collapse in Variational Autoencoders (VAEs) arises when the variational posterior distribution closely matches the prior for a subset of latent variables. This paper presents a simple and intuitive explanation for posterior…

Machine Learning · Computer Science 2019-11-07 James Lucas , George Tucker , Roger Grosse , Mohammad Norouzi

In variational autoencoders (VAEs), the variational posterior often collapses to the prior, known as posterior collapse, which leads to poor representation learning quality. An adjustable hyperparameter beta has been introduced in VAEs to…

Machine Learning · Statistics 2025-07-22 Yuma Ichikawa , Koji Hukushima

The posterior collapse phenomenon in variational autoencoder (VAE), where the variational posterior distribution closely matches the prior distribution, can hinder the quality of the learned latent variables. As a consequence of posterior…

Machine Learning · Statistics 2024-05-14 Hien Dang , Tho Tran , Tan Nguyen , Nhat Ho

Variational autoencoders (VAEs) face a notorious problem wherein the variational posterior often aligns closely with the prior, a phenomenon known as posterior collapse, which hinders the quality of representation learning. To mitigate this…

Machine Learning · Statistics 2023-10-25 Yuma Ichikawa , Koji Hukushima

This work identifies the existence and cause of a type of posterior collapse that frequently occurs in the Bayesian deep learning practice. For a general linear latent variable model that includes linear variational autoencoders as a…

Machine Learning · Computer Science 2022-10-17 Zihao Wang , Liu Ziyin

Variational autoencoders (VAEs), one of the most widely used generative models, are known to suffer from posterior collapse, a phenomenon that reduces the diversity of generated samples. To avoid posterior collapse, many prior works have…

Machine Learning · Computer Science 2025-08-19 Hyunsoo Song , Seungwhan Kim , Seungkyu Lee

The variational autoencoder (VAE) is a popular combination of deep latent variable model and accompanying variational learning technique. By using a neural inference network to approximate the model's posterior on latent variables, VAEs…

Machine Learning · Computer Science 2019-01-30 Junxian He , Daniel Spokoyny , Graham Neubig , Taylor Berg-Kirkpatrick

In narrow asymptotic settings Gaussian VAE models of continuous data have been shown to possess global optima aligned with ground-truth distributions. Even so, it is well known that poor solutions whereby the latent posterior collapses to…

Machine Learning · Computer Science 2019-12-24 Bin Dai , Ziyu Wang , David Wipf

Variational autoencoders (VAEs) are one of the deep generative models that have experienced enormous success over the past decades. However, in practice, they suffer from a problem called posterior collapse, which occurs when the encoder…

Machine Learning · Computer Science 2024-02-06 Yuri Kinoshita , Kenta Oono , Kenji Fukumizu , Yuichi Yoshida , Shin-ichi Maeda

Variational autoencoders model high-dimensional data by positing low-dimensional latent variables that are mapped through a flexible distribution parametrized by a neural network. Unfortunately, variational autoencoders often suffer from…

Machine Learning · Statistics 2023-01-03 Yixin Wang , David M. Blei , John P. Cunningham

Variational autoencoders (VAEs) hold great potential for modelling text, as they could in theory separate high-level semantic and syntactic properties from local regularities of natural language. Practically, however, VAEs with…

Machine Learning · Computer Science 2020-11-11 Teng Long , Yanshuai Cao , Jackie Chi Kit Cheung

Data-driven reduced-order models based on autoencoders generally lack interpretability compared to classical methods such as the proper orthogonal decomposition. More interpretability can be gained by disentangling the latent variables and…

Machine Learning · Computer Science 2025-02-21 Henning Schwarz , Pyei Phyo Lin , Jens-Peter M. Zemke , Thomas Rung

Variational autoencoders (VAEs) suffer from posterior collapse, where the powerful neural networks used for modeling and inference optimize the objective without meaningfully using the latent representation. We introduce inference critics…

Machine Learning · Computer Science 2022-07-21 Sachit Menon , David Blei , Carl Vondrick

Variational autoencoders (VAEs) are a standard framework for inducing latent variable models that have been shown effective in learning text representations as well as in text generation. The key challenge with using VAEs is the {\it…

Machine Learning · Computer Science 2020-05-01 Serhii Havrylov , Ivan Titov

The Variational Autoencoder (VAE) is known to suffer from the phenomenon of \textit{posterior collapse}, where the latent representations generated by the model become independent of the inputs. This leads to degenerated representations of…

Machine Learning · Computer Science 2023-09-12 Fotios Lygerakis , Elmar Rueckert

Variational autoencoders (VAEs) frequently suffer from posterior collapse, where the latent variables become uninformative as the approximate posterior degenerates to the prior. While recent work has characterized collapse as a phase…

Machine Learning · Computer Science 2026-05-19 Zegu Zhang , Jian Zhang

We take steps towards understanding the "posterior collapse (PC)" difficulty in variational autoencoders (VAEs),~i.e. a degenerate optimum in which the latent codes become independent of their corresponding inputs. We rely on calculus of…

Machine Learning · Computer Science 2019-08-01 Octavian-Eugen Ganea , Yashas Annadani , Gary Bécigneul

Being one of the most popular generative framework, variational autoencoders(VAE) are known to suffer from a phenomenon termed posterior collapse, i.e. the latent variational distributions collapse to the prior, especially when a strong…

Machine Learning · Computer Science 2021-03-23 Renfei Tu , Yang Liu , Yongzeng Xue , Cheng Wang , Maozu Guo
‹ Prev 1 2 3 10 Next ›